Overlay Error Measurements
Integrated circuits are very complex devices that include multiple layers. Each layer may include conductive material, isolating material while other layers may include semi-conductive materials. These various materials are arranged in patterns, usually in accordance with the expected functionality of the integrated circuit. The patterns also reflect the manufacturing process of the integrated circuits.
Each layer is formed by a sequence of steps that usually includes depositing a resistive material on a substrate/layer, exposing the resistive material by a photolithographic process, and developing the exposed resistive material to produce a pattern that defines some areas to be later etched.
Ideally, each layer is perfectly aligned to a previously existing layer. Typically, the layers are misaligned, thus a misalignment or overlay error exists between each pair of layers.
Various techniques evolved for observing overlay errors, some using optical instruments and some using scanning electron microscopes. U.S. Pat. No. 6,407,396 of Mih et al., U.S. Pat. No. 6,489,068 of Kye, U.S. Pat. No. 6,463,184 of Gould et al., U.S. Pat. No. 6,589,385 of Minami et al and U.S. Pat. No. 7,842,933 of Shur et al., all being incorporated herein by reference, provide a good indication about the state of art overlay error measurement techniques.
Optical overlay measurements are subjected to various errors such as lens aberrations of the optical system. Mih states that in some cases Atomic Force Microscopy or Scanning Electron Microscopy metrology techniques may be necessary to verify the optical overlay measurement accuracy.
Double patterning is a class of patterning techniques designed to increase the density of circuit features that can be produced on the wafer beyond what the normal limits of a particular lithography scanner. Double patterning can include manufacturing an inspected object that has a first layer and a second layer. The second layer is buried under the first layer. Each layer having features (patterns) that are very close to each other. These features may be eventually etched during the double patterning process.
It is noted that detecting overlay errors after the top layer is etched can be very costly.
Interaction Between Charged Electron Beam and an Inspected Object
Once an electron beam hits an inspected object various interaction processes occur. A detailed description of these processes can be found at “Scanning electron microscopy”. L. Reimer, second edition, 1998, which is incorporated herein by reference.
FIG. 1 illustrates the important interaction processes and various information volumes. An information volume is a space in which interaction processes occur and result in scattering or reflection of electrons that may be eventually detected to provide information about the information volume.
The figure illustrates a primary electron beam 2 that hits an inspected object 9. As a result, secondary electrons 2 and Auger electrons 4 are emitted from a very thin information volume 3 while back scattered electrons (BSE) 6 and X-ray electrons 8 can leave the inspected object from a relatively large information volume 5 that has a depth that may even exceed one micron.
Secondary electrons are easy to detect as their trajectory can be relatively easily changed such that they are directed toward a detector. The trajectory of backscattered electrons is relatively straight and is slightly affected by electrostatic fields.
FIG. 2 illustrates a first type of a prior art multi-perspective SEM 10 that includes multiple detectors. SEM 10 includes an electron gun (not shown) for generating a primary electron beam, as well as multiple control and voltage supply units (not shown), an objective lens 12, in-lens detector 14 and external detectors 16. System 10 also includes deflection coils and a processor (not shown). Such a system is described at U.S. Pat. No. 5,659,172 of Wagner.
In system 10 the primary electron beam is directed through an aperture 18 within the in-lens detector 14 to be focused by the objective lens 12 onto an inspected wafer 20. The primary electron beam interacts with wafer 20 and as a result various types of electrons, such as secondary electrons, back-scattered electrons, Auger electrons and X-ray quanta are reflected or scattered. Secondary electrons can be collected easily and most SEMs mainly detect these secondary electrons.
System 10 is capable of detecting some of the emitted secondary electrons by in-lens detector 14 and by external detectors 16.
Objective lens 12 includes an electrostatic lens and a magnetic lens that introduce an electrostatic field and a magnetic field that leak from the lens towards the wafer. The collection of secondary electrons is highly responsive to the leaked electrostatic field, while it is hardly influenced by the leaked magnetic field.
The leaked electrostatic field attracts low energy secondary electrons and very low energy secondary electrons into the column. A significant part of the very low enemy secondary electrons are directed through the aperture of in-lens detector 14 and are not detected. Low energy secondary electrons are directed towards the in-lens detector 14. High-energy secondary electrons are detected if their initial trajectory is aimed towards one of the detectors.
An effective defect review tool requires both types of detectors in order to capture all types of defects. In-lens detector 14 is usually used for determining a contrast between different materials, and is also useful in voltage contrast mode as well as in HAR mode. HAR mode is used to inspect cavities that are characterized by a High Aspect Ratio (in other words—cavities that are narrow and deep). During HAR mode, the area that surrounds the cavity is usually charged to allow electrons from the lower portion of the cavity to reach the detector. The in-lens detector 14 is also very sensitive to pattern edges. External detectors 16 are much more sensitive to the topography of the wafer. The external detectors are also less susceptible to wafer charging, which is significant when imaging highly resistive layers.
U.S. Pat. No. 6,555,819 of Suzuki et al. (which is incorporated herein by reference) describes a multi-detector SEM having magnetic leakage type objective lens where the magnetic field largely influences the trajectory of emitted secondary electrons. This SEM has various disadvantages, such as not being capable of providing tilted images and is not efficient to provide images from holes of high aspect ratio. Suzuki has a reflector that includes an aperture through which the primary electron beam passes, thus reflected electrons may pass through this aperture and remain un-detected.
There is a need to provide an efficient system and method that facilitate overlay measurements.